© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 193764
    Prev
    Next
    INDEX
    Explanations

    good food, coffee, weather

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    '].
    0.45
     analyzers
    0.44
     guten
    0.42
     хорошие
    0.42
    lamualaikum
    0.41
    有名
    0.40
     bagus
    0.39
     Президента
    0.39
     казіно
    0.39
     አንዳንድ
    0.38
    POSITIVE LOGITS
     wines
    0.43
     healthcare
    0.41
    眠
    0.40
     music
    0.39
     behave
    0.39
     Wines
    0.38
     wine
    0.38
     facilities
    0.37
    िव
    0.37
     supplies
    0.37
    Activations Density 0.015%

    No Known Activations